Original Article

Unsupported Valvuloplasty in Children with Congenital Mitral Valve Anomalies. Late Clinical Results

Gabriel Lorier, Renato A.K. Kalil, Gustavo R. Hoppen, Christiano Barcellos, Nicolas Teleo, Abud Homsi Netto, Javier Gonzalez, Paulo R. L. Prates, Paulo R. Prates, João R. M. Sant'Anna, Ivo A. Nesralla

Porto Alegre, RS - Brazil

OBJECTIVE: To analyze late clinical evolution after surgical treatment of children, with reparative and reconstructive techniques without annular support.

METHODS: We evaluated 21 patients operated upon between 1975 and 1998. Age 4.67±3.44 years; 47.6% girls; mitral insufficiency 57.1% (12 cases), stenosis 28.6% (6 cases), and double lesion 14.3% (3 cases). The perfusion 43.10±9.50min, and ischemia time were 29.40±10.50min. The average clinical follow-up in mitral insufficiency was 41.52±53.61 months. In the stenosis group (4 patients) was 46.39±32.02 months, and in the double lesion group (3 patients), 39.41±37.5 months. The echocardiographic follow-up was in mitral insufficiency 37.17±39.51 months, stenosis 42.61±30.59 months, and in the double lesion 39.41±37.51 months.

RESULTS: Operative mortality was 9.5% (2 cases). No late deaths occurred. In the group with mitral insufficiency, 10 (83.3%) patients were asymptomatic (p=0.04). The majorit y with mild reflux (p=0.002). In the follow-up of the stenosis group, all were in functional class I (NYHA); and the mean transvalve gradient varied between 8 and 12mmHg, average of 10.7mmHg. In the double lesion group, 1 patient was reoperated at 43 months. No endocarditis or thromboembolism were reported.

CONCLUSION: Mitral stenosis repair has worse late results, related to the valve abnormalities and associated lesions. The correction of mitral insufficiency without annular support showed good long-term results.

Key-words: congenital cardiopathy, valvuloplasty, surgery, stenosis, mitral insufficiency

Congenital malformations of the mitral valve are complex lesions resulting from several morphological abnormalities that, generally, involve more than 1 valve component ^1-3 and take place in a population with high prevalence of associated cardiac anomalies ^4-6.

Isolated congenital lesions, such as stenosis or mitral insufficiency, are rare ^1,7,8, representing 1% of the population of congenital cardiopathy patients ⁵.

Congenital mitral insufficiency is extremely uncommon ⁸, being found in childhood associated with other cardiac defects, disorders of conjunctive tissue, and in acquired inflammatory conditions, such as myocarditis, endocarditis, rheumatic fever, Kawasaki's disease, and other colagenosis with vascular impairment ⁸. In congenital mitral stenosis, obstruction to the flow results from morphological anomalies at different levels^1,2,5 and is more frequent than in cardiac insufficiency ².

The purpose of the present study was analysis of late clinical evolution after the surgical treatment of congenital mitral anomalies, with or without associated malformations in children up to 12 years of age treated with reparative and reconstructive techniques without annular support, and to carry out a literature review.

Methods

In the period between 1975 and 1998, 21 children with congenital lesion of the mitral valve were operated on at the Institute of Cardiology of the RS/FUC. The mean age was 4.67±3.44 years; 47.6% were girls and 52.4% were boys. Mitral valve insufficiency was present in 57.1% (12 cases); stenosis was present in 28.6% (6 cases); and the double lesion was present in 14.3% (3 cases) (Fig. 1). The most frequently found morphological lesions in the insufficiency group were annular dilation in 75%, cleft of the posterior leaflet in 33.3% and anterior leaflet cleft in 16.6%. In the stenosis group, parachute valve in 50% and commissura fusion in 33.3% were the most common lesions. In the double lesion group, the most common finding was reduction of leaflet mobility in 100% of patients (Table I). Associated intracardiac malformations were found in 61.9% of the patients, in the stenosis group, 83.3%; in the insufficiency group 41.6%; and in the double lesion group, all patients; the individual description of the different associated intracardiac malformations are presented in Table I. More severe malformations were found in 28.6% of the patients, all with valvular stenosis. Parachute valve was found in 3 cases, all with stenosis; 2 of them were associated with Shone syndrome. Isolated annular dilation without other malformations was present in 2 cases (Table I).

All patients were operated on with median sternotomy and with conventional extracorporeal circulation, using a disposable bubble or membrane oxygenator and moderate hypothermia between 28° and 30°C. Hyperkalemic crystalloid cardioplegia, with cooling of the pericardial cavity with saline solution at 4°C was used for myocardial protection. The approach to the mitral valve was performed by left longitudinal atriotomy. In 2 patients (33.3%), isolated mitral commissurotomy was carried out. In the insufficiency group, annuloplasty was performed with the Wooler's technique in 9 (75%) patients, and the associated procedures in 75% of the cases were 4 chord shortenings, 3 closings of the anterior leaflet cleft, and 2 closings of the posterior leaflet cleft (Table I). The associated cardiac lesions were all corrected after the mitral valvuloplasty.

The mean extracorporeal circulation and aortic clamping times were 43.1±9.5min and 29.4±10.5min.

The postsurgical follow-up was performed by periodic clinical, radiological, and echocardiographic evaluation. The follow-up period varied from 7 days to 15.7 years (mean of 3.8±4.1 years).

Results

The operative mortality in mitral stenosis was 8.9% (2 deaths). One patient, 12 days of age, had hypoplastic mitral valve associated with left ventricular fibroelastosis (Table I).

Overall morbidity was 28.5% (of the total population of patients), being higher in the mitral insufficiency group. Two patients had atelectasis after chest tube withdrawal and pleural effusion due to cardiac failure. One patient in the stenosis group had repetitive respiratory infection. One patient in the double lesion group had infection of the surgical wound.

No episodes of endocarditis or thromboembolism during the follow-up were reported (Table I). The incidence of reoperation was 13% (3 cases). The associated anomalies, as well as the surgical techniques employed, are shown in Table I.

In the insufficiency group, the mean follow-up was 41.52±53.61 months. In the last clinical evaluation, 10 (83%) patients were asymptomatic without medication. One patient was in NYHA functional class I, and 1 patient was in functional class II, both receiving medication (Table I). One patient was operated on again 48 months after surgery, when a new valvuloplasty was performed. The mean echocardiographic follow-up was of 37.17±39.51 months, most of the patients had a light level of valvular regurgitation p=0.002 (Table I).

In the stenosis group (6 cases), 2 deaths and 1 reoperation that evolved to death occurred. Four patients were in functional class I (2 were receiving medication). The echocardiographic follow-up was performed in 4 of these patients (42.61±30.59 months) and showed an average transvalve gradient between 8 and 12mmHg; the mean was 10.7mmHg.

In the double lesion group (3 cases), 1 patient was out-of-control, in whom a slight reflux occurred on the echocardiogram at the time of release from the hospital, and in 1 case, the valve was replaced by a bioprosthesis 43 months after the first surgery. The third patient of the group was in functional class II, and the echocardiography showed stenosis and light insufficiency at 75 months of the surgery.

At the echocardiographic evaluation of the whole sample, with a mean follow-up of 39.89 months, most patients had light lesions (p=0.002) (Table I). In the insufficiency group, 83.3% had absence or light reflux.

The actuarial survival curve in the insufficiency group at 5, 10, and 15 years was of 90% (Fig. 2). The actuarial survival probability free of reoperation in the whole sample (n=21) was 72% at five years and of 46% at 10 and 15 years (Fig. 3). In the insufficiency group, the actuarial survival probability free of reoperation was 86% at 5, 10 and 15 years (Fig. 4).

Discussion

The clinical presentation, as well as the surgical indication, of stenosis or insufficiency depends on the severity of the mitral lesion and the association with intracardiac defects ¹⁰. The surgery in our series was indicated for refractory cardiac failure or important pulmonary hypertension, or both. Preferably, the surgery should not be performed before 6 months of life, according to Carpentier ³. During childhood, before 3 months of age, with inadequately matured collagen, the valvular tissue is particularly friable, making the manipulation difficult ³. In our sample, the mean age of the patients was 5.7 ± 3.3 years in the congenital mitral insufficiency group and 4.0±5.9 years in the congenital mitral stenosis group, without statistical significance (Fig. 1). In Carpentier's series ³, the age of the congenital mitral insufficiency group was 6.1±3.2 years and in the congenital mitral stenosis group 5.1±3.2 years. In the report by Uva and colleagues ¹¹, from Marie-Lannelongue Hospital, the congenital mitral insufficiency group was 7.4±2.7 months old, and the congenital mitral stenosis group was 5.8±3.9 months old. The factors that explain the lower age of the congenital mitral stenosis patients at the time of surgery are 1) the stenotic lesion is less tolerated than the insufficiency; 2) the association with higher and more severe intracardiac anomalies; 3) the abnormalities of the subvalvular apparatus are more frequent in these patients.

Congenital mitral stenosis occurs in 0.6% of the autopsies and in 0.2% and 0.4% of the clinical series ^1,12. In the Ruckman and Van Praagh's ¹³ series of 49 autopsies with congenital mitral stenosis, typical mitral stenosis was found in 49% of the cases, with coarctation of the aorta as the most common associated lesion. The size of the left ventricle was normal in 96% of these patients. Congenital hypoplasia of the mitral valve was the second cause of congenital mitral stenosis (41%), and it was always associated with left ventricular hypoplasia. Supravalvular ring was found in 12% of the cases and the parachute mitral valve in 8%.

Traditionally, the lesions that make the effective and long-lasting repair more difficult are those that have alterations in the subvalvular system with abnormal papillary muscles, including the parachute valve, the hammock valve, and agenesis of the papillary muscles. These lesions determine stenosis with greater frequency ¹⁰ and are associated with the high frequency of complex malformations. Barbero-Marcial and co-workers ¹⁴ achieved good short- and long-term results in 7 patients with mitral parachute valve with stenosis ¹⁴. According to the review presented in Table II, the different publications during the last decade show important progress regarding the results, where it was possible to perform the valvuloplasty in 91.4% of the cases of mitral stenosis with parachute and hammock valve in 50% of the cases.

In the Moore et al. ¹² series, the typical mitral hypoplasia with symmetric papillary muscles was the first cause of congenital mitral stenosis (52% of the cases), followed by the supravalvular ring (in 20%), double orifice (in 11%), mitral hypoplasia with asymmetry of the papillary muscles (in 8%), and parachute-shaped mitral valve (in 8%). Contrary to this series, Embrey and Behrendt ⁵ state that the ring is rarely so small that stenosis results, unless left ventricular hypoplasia is present. According to these authors, chordal malformation is the most common cause of stenosis ⁵.

In a series of 50 patients, Fuzzelier et al. ¹⁵ reported commissura fusion as the most common cause of congenital mitral stenosis with normal papillary muscles (in 17 patients). With abnormal papillary muscles, the most common was the hammock mitral valve in 11 patients. In our series, 50% of the cases of congenital mitral stenosis had abnormal papillary muscles; as the most prevailing cause, the parachute mitral valve in 3 patients, 2 of them associated with Shone syndrome, with possible recovery (Table I); 38% typical mitral stenosis and 12.5% hypoplasic mitral valve (Table I). Left ventricular hypoplasia was present in 50% of the cases of congenital mitral stenosis, associated in 50% with left ventricular fibroelastosis.

According to McGiffin ², congenital mitral insufficiency is less frequent than congenital mitral stenosis. In our series, we observed 12 cases of insufficiency and 6 cases of stenosis. The most frequent cause in congenital mitral stenosis is annular dilation ^2,16; this finding was confirmed in our sample. Seventy-five percent of the patients in the congenital mitral insufficiency group had this malformation, and in 25% this was the only malformation.

Congenital or acquired abnormalities of the mitral valve in children can be surgically managed with mechanical or biological prosthesis or valvuloplasty ^4,17. Valvular replacement is accompanied by a high mortality rate and by anticoagulation problems and the impossibility of annular growth, leading inevitably to re-operation ^2,4,17-22.

The use of a rigid or flexible ring as a prerequisite for efficacy and durability of annular remodeling ²³ is currently being questioned in the adult ^24,25 and a tendency to reduce its use has been observed ^26,27.

In children and adolescents, the prosthetic rings must be avoided for they do not allow a normal annular growth ⁵, in addition to being a risk factor for distortions of the ventricular cavity and for contributing to the obstruction of the left ventricular outlet ^18,28. Without adding the fact that the annular segment of mitral-aortic continuity is capable of contracting and relaxing during the cardiac cycle in the left ventricular outlet ^25,26,28, because it is the only annular portion that does not dilate, which leads to the conclusion that at the described level no prosthetic structure should exist.

The long-term stability of mitral valve repair within the concept of remodeling annuloplasty does not imply the use of a rigid or flexible ring to decrease the antero-posterior ring diameter. Since 1975, we have believed that the stability in Wooler's technique ⁹ of unsupported annuloplasty is contained in the anchorage of the right and left fibrous trigones (Fig. 5), keeping the normal variable anatomical relationships with the aortic valve and the left ventricular outlet, and the annular segment of mitral-aortic continuity with the capacity of contraction and relaxation during the cardiac cycle ²⁸.

At the time of performing Wooler's technique, it is of essential importance to long-term repair stability and to avoid failure, that the points undergo through the fibrous trigones, reminding their position related with the anterior cuspid ²⁹. This technique has a late result comparable to more complex support technqiues ^30-34, but with a lower incidence of technical repair failure. The reported incidence of repair failure due to technical problems with the prosthetic ring in the adult is varied (Aharon ¹⁰ - 2.9%; Cosgrove ³⁴ - 3.2%; Deloche ³⁵ - 4.3%; Cosgrove ²⁴ - 3.3%).

Regarding intraoperative deaths, patients with congenital mitral stenosis who were in NYHA functional class IV had in our series 100% mortality and in Kirklin's series 50% mortality ^36,37; both patients had endocardial fibroelastosis. The 12-day-old patient had left ventricular hypoplasia, and the 7-month-old had Shone syndrome. Recent studies carried out by Ni and colleagues ³⁸ suggest a viral infection as the cause of endocardial fibroelastosis, supporting the hypothesis that this disease is a sequela of viral myocarditis, particularly due to the mumps virus.

The mitral corrections were performed by longitudinal atriotomy of the left atrium's right wall, without using other approaches ^14,39-43.

The comparative results among the studies found in the literature about congenital mitral insufficiency, presented in chart I, corroborate the good long-term results of the unsupported annuloplasty in this population of patients, also found in our series. However, the population of patients with congenital mitral stenosis has a variety of mitral valve malformations, frequent associations with cardiac malformations, and variable age, making it difficult to obtain generalized conclusions, according to chart II.

In conclusion, mitral valvuloplasty in isolated congenital lesions or in association with other cardiac malformations has good long-term results. Failure to repair stenosis and double lesions is due to the complexity of malformations detected.

In the case of congenital mitral insufficiency, the unsupported mitral valvuloplasty with Wooler's technique had a low surgical risk and good long-term results. The use of prosthetic rings in these patients is considered unnecessary.

Instituto de Cardiologia do Rio Grande do Sul - Fundação Universitária de Cardiologia

Mailing address: Gabriel Lorier – IC/FUC - Av. Princesa Isabel, 395 - 90620-001 - Porto Alegre, RS – E-mail: pesquisa@cardnet.tche.br